Pilley, Tania; (1999) A psychometric study of the child attachment interview: Reliability and validity. Doctoral thesis , UNSPECIFIED.

Text A_psychometric_study_of_the_ch.pdf Download (6MB)

Abstract

The Child Attachment Interview (CAI) is being developed to explore children's mental representations of attachment as manifested through the content of their language and behaviour. The attachment measure intends to bridge the gap that currently exists for children in middle childhood (aged 8-12). This study has explored the psychometric properties of the CAI in an effort to determine its reliability and validity. The sample consisted of 15 mothers who had 30 children between them. The mothers were administered the Adult Attachment Interview (AAI) and the thirty children were administered the CAI. A re-administration of the CAI was carried out on 28 of these children approximately two months later in order to evaluate test-rest reliability. The reliability of the CAI classifications proved to be high over time, suggesting a stability of responses across time. Inter-rater reliability was also established at a high level and the internal test construction was found to be very satisfactory. To examine the CAI's discriminant validity, the Wechsler Intelligence Scale for Children -Third Edition (WISC- III UK), and the Clinical Evaluation of Language Fundamentals-Revised (CELF-R UK) were administered. Parents also completed the Child Behaviour Check List (CBCL) to measure for clinical morbidity. Concurrent validity was assessed by observing the correlation between the Separation Anxiety Test (SAT) and the CAI. Predictive validity was assessed through the mother's interview transcripts from the AAI. The CAI classifications were found to be independent of demographic variables such as, gender, age, social economic class, one or two parent households and ethnicity. The classifications were also independent of discriminant variables such as, intelligence, language ability and clinical morbidity. Only a weak correlation was found between the SAT and the CAI. However, the reliability and validity of the SAT is questionable. Clear evidence of the predictive validity of the CAI was demonstrated by a 68% concordance rate between mother and child interviews, which is consistent with previous research findings. In summary, the evidence for the psychometric properties of the CAI seems encouraging. However, concurrent validity needs to be established.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item